Great work. Such a complicated Cattleya stew. You have described the role natural selection seems to have played and then humans chopped and mixed things further. I think I am just going to focus on whether I have plants that grow and bloom well and whether I think the flowers are attractive! A long way from where I started.
Proposal to evaluate genetic tools to evaluate Cattleya species
- Thread starter geoffsharris
- Start date
Help Support Slippertalk Orchid Forum:
treefrog
Well-Known Member
This is not something I enjoy doing but I feel it’s important to warn people that the ‘evolution’ of Cattleya as presented in this post is purely speculative. None of the genetic data publicly available can be used to assess introgression or hybridization history (mitochondria and chloroplast do not recombine!) or any of the demographic history presented here. Population genomic is an extremely intricate field of research with methods originating from genetic, phylogenetic, mathematic, ecology, demography and this using complex modelling, statistics, and now deep learning . Population genomic not a field you simply become proficient by looking at sequences…unfortunately.
Last edited:
- Joined
- May 21, 2014
- Messages
- 90
- Reaction score
- 126
I'm sure Mathieu is genuinely just trying to be helpful in warning everyone, but I think this is pretty misguided. For those that are genuinely curious to think more deeply on this, are unafraid of being exposed to speculative hypotheses (that could be wrong) and are prepared to ignore his warning, read on.
The information as presented is entirely speculative as is clear from the post. It is my opinion and nothing more. That opinion is informed by 40 years of growing these plants, meticulously mapping out where these plants do and don't live, looking at the morphology of the plants both macroscopically and microscopically, reading prodigiously the original descriptions, papers speculating on the genetic origins of Cattleya and other orchids and looking up as many online herbarium sheets as exist for these species. It is further reinforced by visiting a number of habitats, growing multiple examples of every one of these species except 2, speaking with exceptionally knowledgable growers and experts, examining what is known of natural hybrids and looking at man made ones that suggest some of these ideas have merit - just go look at what luteola does when crossed with a large purple flowered cattleya species if you think my idea on dowaia/aurea/rex origins are nonsense or look at mossiae x lawrenceana and how much this looks a lot like a proto lueddemanniana.
Nothing in this set of speculative hypotheses is out of line with what has been more or less previously proposed by Carl Withner in The Cattleya and their Relatives or Cassio Van den Berg in the various papers he has written.
http://dx.doi.org/10.11646/phytotaxa.186.2.2
https://www.researchgate.net/public...anscribe_Spacers_ITS_of_nuclear_ribosomal_DNA
There is a very nice description of the phenomena of hybrid speciation that suggests this is a relatively common source of speciation within orchids by Casio Van Den Berg in Renziana #4 Cattleyas
https://koeltz.com/en/zeitschrift-d...on-vol-4-cattleya-2014-illus-98-p4to-paper-bd
For the intrepid, there is data in the public domain for both chloroplast gene sequences and internal transcribed spacer 1, 5.8S ribosomal RNA, and internal transcribed spacer 2 which is genomic (ITS). Mathieu is correct that chloroplast genes are maternally transmitted and are essentially haploid. ITS is genomic. Mutations in the chloroplast genes are super useful for clustering groups of species that are more closely related vs. distantly related, but very poor for identifying differences between closely related species. The most obvious evidence for a species of hybrid origin is that is has an ITS gene that clusters with one set of plants and one or more chloroplast genes that cluster with another separated group. If this is further supported by intermediate characteristics in growth and flowering and a plausible geographic relationship, it's pretty likely to be true. This is evident in a bunch of the rupiculous Cattleyas, Cattleya aclandiae, schilleriana and others in the unifoliate and bifoliate cattleya groups.
You do need to be curious about how genetics work, be willing to download some software and data, understand the limitations of the data sets and how various approaches to statistical relationships in building phylogenic trees don't work all that well in plants that have reticulated genetic histories, etc. but the idea that you need to be some sort of genius with millions of dollars in grants and a PhD in genetics to have an opinion on this is wrong.
The whole point of this thread was to find simple ways to identify plants as a specific species rather than a first or second generation hybrid. There are a litany of examples such as Cattleya mossiae 'Willowbrook', Cattleya walkeriana 'Pendentive' and 'Kenny' and Cattleya eldorado 'Mt. Ito' that are known or suspected hybrids where this would be useful. If you are curious, look a the paper - DNA barcode regions for differentiating Cattleya walkeriana and C. loddigesii Doi: 10.4025/actascibiolsci.v39i1.33024
That shows a single base in the rPOC1 gene is sufficient to differentiate between lodgigesii and walkeriana thereby accomplishing the goal of having a simple diagnostic tool for plants like 'Kenny' and 'Pendentive'. I am positive the difference between mossiae and lueddemanniana could be similarly reduced to a very simple single gene test for telling apart x gravesiana, but not at all sure for more introgressed clones. As for the rest of the unifoliate Cattleyas, do I think this could be done to tell apart trianae from schroderae and candida to know if you have a hybrid or species plant? No, I don't think so as I believe that the genetic histories of these plants is one of crashing together and separating over time such that they have gone back and forth so much that today they are separated in space, but genetically so inter related that the variability within each population is likely overlapping with each of the other species for many genes. I would suggest that the species concept in orchids is at it's limit in any number of species where there are intergrades at the margins - walkeriana and nobilor, loddigesii and harrisoniana, etc. So, in specific cases do I think we could make a similar diagnostic tool for specific conundrums as was done for lodgigesii and walkeriana, yes. However, my point is that these would not be generalizable to asking more complicated questions of is my plant this or that species or a hybrid based on an informed speculation that the genetic histories for many of the species is likely to be so much more complicated at the population level such that simple yes/no approaches would be ineffective in most cases.
Informed speculation and hypothesis formation is at the root of any effort to further understand how the world works and informed speculation essentially describes every published orchid book or article ever written. I welcome anyone to use their knowledge to help propel these ideas forward. Feel free to critic the hypotheses presented, add to them or beat them up to make them better.
By the way, my suggested speculations are entirely consistent with the following genetic tree based on ITS data. From Rivera-Jiménez 2017 - Doi: 10.4025/actascibiolsci.v39i1.33024
The information as presented is entirely speculative as is clear from the post. It is my opinion and nothing more. That opinion is informed by 40 years of growing these plants, meticulously mapping out where these plants do and don't live, looking at the morphology of the plants both macroscopically and microscopically, reading prodigiously the original descriptions, papers speculating on the genetic origins of Cattleya and other orchids and looking up as many online herbarium sheets as exist for these species. It is further reinforced by visiting a number of habitats, growing multiple examples of every one of these species except 2, speaking with exceptionally knowledgable growers and experts, examining what is known of natural hybrids and looking at man made ones that suggest some of these ideas have merit - just go look at what luteola does when crossed with a large purple flowered cattleya species if you think my idea on dowaia/aurea/rex origins are nonsense or look at mossiae x lawrenceana and how much this looks a lot like a proto lueddemanniana.
Nothing in this set of speculative hypotheses is out of line with what has been more or less previously proposed by Carl Withner in The Cattleya and their Relatives or Cassio Van den Berg in the various papers he has written.
http://dx.doi.org/10.11646/phytotaxa.186.2.2
https://www.researchgate.net/public...anscribe_Spacers_ITS_of_nuclear_ribosomal_DNA
There is a very nice description of the phenomena of hybrid speciation that suggests this is a relatively common source of speciation within orchids by Casio Van Den Berg in Renziana #4 Cattleyas
https://koeltz.com/en/zeitschrift-d...on-vol-4-cattleya-2014-illus-98-p4to-paper-bd
For the intrepid, there is data in the public domain for both chloroplast gene sequences and internal transcribed spacer 1, 5.8S ribosomal RNA, and internal transcribed spacer 2 which is genomic (ITS). Mathieu is correct that chloroplast genes are maternally transmitted and are essentially haploid. ITS is genomic. Mutations in the chloroplast genes are super useful for clustering groups of species that are more closely related vs. distantly related, but very poor for identifying differences between closely related species. The most obvious evidence for a species of hybrid origin is that is has an ITS gene that clusters with one set of plants and one or more chloroplast genes that cluster with another separated group. If this is further supported by intermediate characteristics in growth and flowering and a plausible geographic relationship, it's pretty likely to be true. This is evident in a bunch of the rupiculous Cattleyas, Cattleya aclandiae, schilleriana and others in the unifoliate and bifoliate cattleya groups.
You do need to be curious about how genetics work, be willing to download some software and data, understand the limitations of the data sets and how various approaches to statistical relationships in building phylogenic trees don't work all that well in plants that have reticulated genetic histories, etc. but the idea that you need to be some sort of genius with millions of dollars in grants and a PhD in genetics to have an opinion on this is wrong.
The whole point of this thread was to find simple ways to identify plants as a specific species rather than a first or second generation hybrid. There are a litany of examples such as Cattleya mossiae 'Willowbrook', Cattleya walkeriana 'Pendentive' and 'Kenny' and Cattleya eldorado 'Mt. Ito' that are known or suspected hybrids where this would be useful. If you are curious, look a the paper - DNA barcode regions for differentiating Cattleya walkeriana and C. loddigesii Doi: 10.4025/actascibiolsci.v39i1.33024
That shows a single base in the rPOC1 gene is sufficient to differentiate between lodgigesii and walkeriana thereby accomplishing the goal of having a simple diagnostic tool for plants like 'Kenny' and 'Pendentive'. I am positive the difference between mossiae and lueddemanniana could be similarly reduced to a very simple single gene test for telling apart x gravesiana, but not at all sure for more introgressed clones. As for the rest of the unifoliate Cattleyas, do I think this could be done to tell apart trianae from schroderae and candida to know if you have a hybrid or species plant? No, I don't think so as I believe that the genetic histories of these plants is one of crashing together and separating over time such that they have gone back and forth so much that today they are separated in space, but genetically so inter related that the variability within each population is likely overlapping with each of the other species for many genes. I would suggest that the species concept in orchids is at it's limit in any number of species where there are intergrades at the margins - walkeriana and nobilor, loddigesii and harrisoniana, etc. So, in specific cases do I think we could make a similar diagnostic tool for specific conundrums as was done for lodgigesii and walkeriana, yes. However, my point is that these would not be generalizable to asking more complicated questions of is my plant this or that species or a hybrid based on an informed speculation that the genetic histories for many of the species is likely to be so much more complicated at the population level such that simple yes/no approaches would be ineffective in most cases.
Informed speculation and hypothesis formation is at the root of any effort to further understand how the world works and informed speculation essentially describes every published orchid book or article ever written. I welcome anyone to use their knowledge to help propel these ideas forward. Feel free to critic the hypotheses presented, add to them or beat them up to make them better.
By the way, my suggested speculations are entirely consistent with the following genetic tree based on ITS data. From Rivera-Jiménez 2017 - Doi: 10.4025/actascibiolsci.v39i1.33024
Last edited:
Wow
